e46 M3 suspension setup, or how to not downgrade your car with suspension mods

[cobra]

Someone please correct me if I'm wrong about any of this.

AFAIK, "flat ride" ride frequencies increase the platform's inherent pitch stability when dampers aren't factored in. That means you don't need as much damping to get a certain level of pitch control, which might simplify damper tuning and possibly allow the car to have more compliance and/or quicker reactions.

IOW, it's an advantage – not a guarantee.

It's also a constraint that might require other tradeoffs, e.g. limiting front spring rates, requiring a chassis-destroyingly stiff front ARB, etc. So, right off the bat, there's room for people to disagree on the "best" setup.

Then there's damping, which is a black art; it introduces more variables and could swing things either way. A non-flat-ride car with excellent damping could be better than a flat-ride car with less-than-excellent damping.

And when you start getting into what's faster or feels better, that introduces human factors, which makes things another order of magnitude more complicated.

AFAICT, this is why most sophisticated suspension tuners treat flat ride as an advantage to be weighed against other factors. I'm pretty sure FCM is unique in prioritizing flat ride and working everything else around it.

So, yeah, no surprise different setups work differently for different people.

FWIW

If you look into the origins of "flat ride", it dates back to the early 1900's when cars were quite a bit different. Nowadays, people look at a car that comes from the factory with stiffer rear spring rates and assume it's that way for some sort of performance advantage, but likely it's to account for the payload rating and still maintain a reasonable chassis attitude and handling. I don't think pickup trucks have stiff rear springs for a handling benefit. The E46 has to account for 5 passengers and a pretty large trunk full of stuff, hence the progressive spring to try to make it work in a variety of situations without looking like one of those trucks that are lifted in front and lowered in the rear

I don't buy into the concept because there is rarely a situation where you're driving in a straight line at a specific speed and hit a certain type of bump at a perpendicular angle. Bumps are random and a properly set up damper will quell oscillations quickly anyway. Also why corner balancing a street driven car is probably a waste of time for 99.9% of people.

You're exactly right, maybe there are advantages in some cases, but you can't just put stiff rear and soft front springs and expect your car to feel comfortable and handle well. It's the whole combination of ride heights, bump stops, spring rates, damper tuning, roll bars, and tires. I don't have the magic formula and there are probably multiple right answers that can work great.

There is also the personal preference aspect. I don't mind feeling bumps in the road, it lets me know what's going on and gives me a degree of confidence. What I don't like is when I feel bumps or vibrations out of sync with what I am expecting to feel. Some people prefer a magic carpet ride and want to float along with zero inputs to the driver until the suspension bottoms out. People also tolerate body roll differently, I think I like a more flat chassis from a confidence perspective even though some body roll might be faster.

[LSB4Me]

Update and important points of note regarding a more comfortable Ohlins setup for the street (I'm calling it the Strohlins Kit):

First, the front spring on the E9x Ohlins RT kit is 60 n/mm (approx. 342lb) and is the same shape, length, and design as the 70 n/mm front spring on the E46M3 RT kit. I am investigating as to whether the E9x spring can be swapped onto the E46M3 strut to achieve a lower front rate without requiring different sized, aftermarket springs (and camber plates).

Second, the upper spring perch on the F80 Ohlins RT kit is made to accommodate a 65mm spring. I am investigating as to whether the F80 perch can be mounted to the E46M3 stock upper strut mount to allow us to change to aftermarket spring sizes without necessitating the addition of camber plates.

I hope to have confirmation on both points this time next week and will revert.

[duracellttu]

I think that is where people get lost. A 300 lb spring is nearly twice as stiff as stock (155 lb/in). That's a huge increase!

I totally recognize that stock springs are much softer when compared to the rates we are discussing. When it comes to personal preference of feel and stability on the road I believe 400# is a good place to start.

[Bry5on]

If you look into the origins of "flat ride", it dates back to the early 1900's when cars were quite a bit different. Nowadays, people look at a car that comes from the factory with stiffer rear spring rates and assume it's that way for some sort of performance advantage, but likely it's to account for the payload rating and still maintain a reasonable chassis attitude and handling. I don't think pickup trucks have stiff rear springs for a handling benefit. The E46 has to account for 5 passengers and a pretty large trunk full of stuff, hence the progressive spring to try to make it work in a variety of situations without looking like one of those trucks that are lifted in front and lowered in the rear

I don't buy into the concept because there is rarely a situation where you're driving in a straight line at a specific speed and hit a certain type of bump at a perpendicular angle. Bumps are random and a properly set up damper will quell oscillations quickly anyway. Also why corner balancing a street driven car is probably a waste of time for 99.9% of people.

You're exactly right, maybe there are advantages in some cases, but you can't just put stiff rear and soft front springs and expect your car to feel comfortable and handle well. It's the whole combination of ride heights, bump stops, spring rates, damper tuning, roll bars, and tires. I don't have the magic formula and there are probably multiple right answers that can work great.

There is also the personal preference aspect. I don't mind feeling bumps in the road, it lets me know what's going on and gives me a degree of confidence. What I don't like is when I feel bumps or vibrations out of sync with what I am expecting to feel. Some people prefer a magic carpet ride and want to float along with zero inputs to the driver until the suspension bottoms out. People also tolerate body roll differently, I think I like a more flat chassis from a confidence perspective even though some body roll might be faster.

Gotta say I’m surprised by your statement that pitch balance only applies to one speed. I thought we took the same ground vehicle courses and learned from the same books (and experience! I was playing with springs and roll centers on my e31 to lock these concepts with reality at the time)
The pitch balance applies at a huge range of speeds, with maximum return at a single speed. You can build a pretty simple excel model to prove this, which is what I did for my car and various spring rate combinations. Basically just build a half car model and plot the pitch angle of the car for a step change in response and you can dial in whatever critical damping decay you want. Bumps are effectively just sums of step changes so you can make the model more advanced if you’d like. I know you know all this, which is why I’m still surprised!

Also, to IamFODI above - the damper doesn’t change the spring natural frequency, so this math applies both with and without dampers. There’s a caveat here! If your springs are sufficiently stiff, and your wheelbase sufficiently long, your springs might settle so fast that pitch balance is irrelevant! This can easily be true with aero race cars, but certainly not for those of our e46s which are street driven and the ride frequencies we’re discussing generally here ~2Hz and below.

The OEMs go through all of this balancing act, but even the fast ones like modern GT Porsches with 2Hz ride rates setting Nurburgring records seem to end up with some form of pitch balance. Always down to personal preference at the end of the day, but I prefer to use the math to help me the same way big factory teams do.

[Bry5on]

I totally recognize that stock springs are much softer when compared to the rates we are discussing. When it comes to personal preference of feel and stability on the road I believe 400# is a good place to start.

You’re giving up ultimate grip on a road car by going up in rates I think most people miss this point. As long as you’re not in the bump stops, you want softer if you want grippier. You might prefer the slower cornering car that transitions faster, but that’s the trade your making, to make it clear!

[cobra]

Gotta say I’m surprised by your statement that pitch balance only applies to one speed. I thought we took the same ground vehicle courses and learned from the same books (and experience! I was playing with springs and roll centers on my e31 to lock these concepts with reality at the time)
The pitch balance applies at a huge range of speeds, with maximum return at a single speed. You can build a pretty simple excel model to prove this, which is what I did for my car and various spring rate combinations. Basically just build a half car model and plot the pitch angle of the car for a step change in response and you can dial in whatever critical damping decay you want. Bumps are effectively just sums of step changes so you can make the model more advanced if you’d like. I know you know all this, which is why I’m still surprised!

Also, to IamFODI above - the damper doesn’t change the spring natural frequency, so this math applies both with and without dampers. There’s a caveat here! If your springs are sufficiently stiff, and your wheelbase sufficiently long, your springs might settle so fast that pitch balance is irrelevant! This can easily be true with aero race cars, but certainly not for those of our e46s which are street driven and the ride frequencies we’re discussing generally here ~2Hz and below.

The OEMs go through all of this balancing act, but even the fast ones like modern GT Porsches with 2Hz ride rates setting Nurburgring records seem to end up with some form of pitch balance. Always down to personal preference at the end of the day, but I prefer to use the math to help me the same way big factory teams do.

I never took that class...

I'm not saying pitch balance isn't a thing, I'm just saying you can't put on stiffer rear springs relative to the front and suddenly expect a magical ride. Some are putting a lot of focus on "flat ride" springs when the entire system needs to be considered.

[Tbonem3]

You’re giving up ultimate grip on a road car by going up in rates I think most people miss this point. As long as you’re not in the bump stops, you want softer if you want grippier. You might prefer the slower cornering car that transitions faster, but that’s the trade your making, to make it clear!

I think I'd go for slightly narrower tires before stiffer springs if I felt responsiveness was lacking.

[Bry5on]

I think I'd go for slightly narrower tires before stiffer springs if I felt responsiveness was lacking.

Good point, and keep a higher offset! Scrub radius definitely influences that responsiveness feeling in my experience. The M3 seems to have a little over 10mm of positive scrub radius from factory if I did my measurements right. An et35 wheel doubles scrub to ~20mm and et25 triples it! Those are big changes. An et45 17x9 is my dream wheel.

Since we’re talking about wheels/tires, running a shorter sidewall tire is the easiest way to lower the CG of the car without upsetting geometries. It’ll come with some unsightly fender gap under M3 fenders though. 245/40/17 under non-m fenders here, 1/2” lower CG than a stock M3.

[Tbonem3]

Agreed, I'd always feel it when I added more than about a 3-5mm spacer. But you won't find me fitting ET45 wheels 😅

[K-Dawg]

My journey for a OEM+ comfortable street setup continues. As you mentioned my attempt at using 350/700 on the TCK SA didn't work out. After reaching out to TC he told me that the SA damper limit was a 700# spring, but didn't have enough compression to be good and recommended a DA. I didn't want to go that route so I opted for a FCM custom setup. Even with those dampers I wasn't getting a comfortable ride with a 700#. I've attempted to get flat ride to work on my street car with little success. Maybe it's the bad LA roads, not sure. Moving back to a 350/500 setup completely eliminated the oscillating rear, but I felt the front end needed to be stiffer.

That being said, my current setup is 400/600 but I don't have much seat time since I'm doing some new engine and tune upgrades. TC's recommendation for “great balance” is 400/500 on a street car. After all the experimenting, I agree that most would be happy with that setup. If you want stiffer, TC recommended 400/700. TBoneM3 has a lot of good suspension info on his journal page, which I've been following. He has gone down this same path we are discussing.

IMO flat ride should only be considered for track cars. It has proven results and many members really like it. Some might argue otherwise, but for me it wasn't providing the street solution I was looking for. Hope this helps.

Flat ride only for track cars? Why have many (most?) suspension designers historically used it on street cars? The E46 M3 is intended to be driven on the street, and has flat ride from the factory. Although, it seems manufacturers have gone to a more pitchy, overdamped setup for some reason. More "sporty" feel, don't care, don't know any better, some actual reason?

What do you mean by "oscillating rear"? It doesn't make sense that you would have more movement with stiffer springs.

Flat ride theory makes sense to me. While a suspension designer may be able to use brute force damping to mask a naturally unsettled car, it seems more efficient for the car to be naturally settled without added damping, then apply a small amount of damping to fine tune things as necessary. The inertia of a car bouncing around back and forth is not conducive to grip or comfort, and using damping to cover this seems like a band-aid.

I can say this... In the ownership of my M3, as I've gone from 550/600 lb/in F/R spring rates, to 300/600 and now 155/375-750 (Eibach), the car has become far more pleasant to drive.

I'm intrigued by the quest for desirable handling characteristics while still maintaining comfort. In my mind, a "perfect" suspension would allow the car body to "float" (have essentially no vertical movement) as it goes down the road, and at the same time the tires are maintaining a steady or constant force on the pavement as they roll down the road. I suppose this would be a numb driving experience, so it would also have to have perfect road feel.

Like lots of other things in life, the more I've learned about suspension design and theory, the more I realize I don't know (I don't know much). I've also come to realize that most people have absolutely no clue what they are talking about when it comes to suspension systems. Automobile suspensions are far more complex than they appear on the surface. Most of us don't have the opportunity to try a significant number of different spring/damper setups over the ownership of our car, let alone test multiple setups back to back in a somewhat controlled environment. And even if we did have that kind of opportunity, most of us would not be able to tell what is actually going on with the suspension from the driver's seat, or accurately determine what was changed in a blind test.

Good discussion!

[Tbonem3]

Regarding Duracell's experience, my immediate thought was that the damper is not controlling (rebound) the stiff 700lb spring well enough, which makes it feel bouncy.

Try a stout damper like MCS! My 628lbs feel so perfect. No bounciness, just firm. I only needed to slow the rebound by a few clicks on the mcs. Even on full soft it had almost slow enough rebound. Dampers make a huge difference! Feels better than my B6s with 550lb springs.

[DJAM3]

That was also my thought regarding the "oscillating rear". Sounds like it was underdamped.

[cobra]

Edit: I redid my measurements tonight more carefully. I did them multiple times to account for measuring errors - that's what Rear 1, 2, 3, etc means. The results were pretty dang consistent, and most importantly the rear is different than the values I was referencing which were 0.70 and 1.05.

Front = 0.94
Rear spring = 0.63
Rear shock = 1.14

Based on this...

-Ohlins 400/630 spring rates yield 2.1Hz and 1.82Hz, or 86% R/F ratio.
-With a 630lb rear spring and 300lb front spring, both will be at 1.82Hz.

I would really love for someone else to do these measurements and make sure we're all using the right numbers. Small differences in spring ratio has a big effect on frequency...

[Slideways]

Comparing the stock spring rates to some of the considerably higher ones mentioned in this thread made me think of this video. A. Biermann, BMW's VP of engineering, said that the E46 was setup pretty stiff from factory and some people complained on the test rides lol. Starts at 3:58

[mrgizmo04]

Make sure to watch a lot of Shailh's videos on his channel, there is a lot to learn.

Here is a demo of gas shock pressure effect https://youtu.be/DOK-BvIORVQ

I will repeat what I have said before. When I did my dyno session with him on Ohlins strut/shock at his shop in Redwood City, we discovered that there is way too much pressure in them and the valving was way tight. Force vs velocity curves (he printed me a copy but I misplaced it somewhere) were close to straight lines, so he recommended to run both fronts and rears at full soft irrespective of spring rates. Not ideal but it would help counteract the initial pressure and smooth out curves. I run them at soft street/autox/track without any adjustments.

Flat ride is a simple concept, where hang time needs to be introduced in the front via lower frequency/softer spring (to react slower) so that once the front rolls over that bump it gives enough time for the wheelbase to travel over it and allow the rear to hit the same bump so that both front and rear settle back down "together". Depends on the wheelbase, weight distribution, etc. Running high frequency/stiff fronts makes the car pitchy, where the front is disconnected from the rear. So when you run over thebump, the front reacts and settles quickly, then the rear runs over the bump and settles independently. That is a much less comfortable/bouncy ride. Vs having front and rear settle at the same time.

If you don't need the car to transition fast like on an autox course, you can go much softer all the way around as long as you don't bottom out.

Just remember that springs impact front to rear comfort (flat ride vs pitch) while also the roll side to side. Swaybars only impact roll. You need all components to work together to get proper balance.

Sent from my SM-G950U using Tapatalk